|
--- |
|
title: README |
|
emoji: 🐨 |
|
colorFrom: blue |
|
colorTo: purple |
|
sdk: static |
|
pinned: false |
|
--- |
|
|
|
For easier cloud AI development, the recipe below has security, performance, lifecycle management and scalability superpowers to make it easy. |
|
|
|
Preparation and model teaching: |
|
|
|
0. Use Huggingface as Fastest way to document and plan with fast code testing. |
|
- https://huggingface.co/awacke1 |
|
1. Create a video playlist for Youtube that shows short videos on pattern implementation: |
|
- https://www.youtube.com/playlist?list=PLHgX2IExbFosXrdx5gTJO33VoDH9bXlFJ |
|
- Keep up with the latest in AI with Wes: https://www.youtube.com/watch?v=SQup5CeRGCM |
|
2. Use Azure Free to set up initial patterns and gain familiarity |
|
- https://portal.azure.com/#view/Microsoft_Azure_Billing/FreeServicesBlade |
|
3. Use an Azure pre-flight checklist to plan your resource selection which we will eventually get down to code. |
|
- https://portal.azure.com/?quickstart=true#view/Microsoft_Azure_Resources/QuickstartCenterBlade |
|
4. New Azure subscription using an Account to explore $200 with resources free, Bookmark opening resources for starters. |
|
- My links are usually Subscription, Resource Group, ACR, ACA Apps. |
|
- For security tests, make it bulletproof - links with email and subscription GUID using self defense tests. |
|
- Sub: https://portal.azure.com/#@AaronCWackergmail.onmicrosoft.com/resource/subscriptions/003fba60-5b3f-48f4-ab36-3ed11bc40816/overview |
|
- RG: https://portal.azure.com/#@AaronCWackergmail.onmicrosoft.com/resource/subscriptions/003fba60-5b3f-48f4-ab36-3ed11bc40816 |
|
- ACR: https://portal.azure.com/#@AaronCWackergmail.onmicrosoft.com/resource/subscriptions/003fba60-5b3f-48f4-ab36-3ed11bc40816/resources |
|
- Apps: https://portal.azure.com/#@AaronCWackergmail.onmicrosoft.com/resource/subscriptions/003fba60-5b3f-48f4-ab36-3ed11bc40816/resourceGroups/acae/providers/Microsoft.App/managedEnvironments/acae |
|
5. Github integration for subscription, github actions, and accelerators |
|
- Accelerator: https://github.com/Azure/aca-landing-zone-accelerator |
|
- LZA: https://github.com/topics/lza |
|
6. Git with one easy example per type of app and secure access pattern. Do easiest first, refactor to easiest. |
|
- HTML5: Depth anything with Transformers.js: https://huggingface.co/spaces/awacke1/3D-Depth-Model-From-Image |
|
- Gradio: Models, Datasets, Spaces, Search Hub: https://huggingface.co/spaces/awacke1/Models-Datasets-Spaces-Search-Hub |
|
- Streamlit: Demonstrate transcript insights using AI libraries: https://huggingface.co/spaces/awacke1/Transcript-EDA-NLTK |
|
7. Get your mobile device for MFA, SMS (IPad, Android, iPhone to demo) |
|
8. Get your top LLM's to help with AI Pair Programming. Refactor model approach to use securely and internally. |
|
- GPT: https://chatgpt.com/ |
|
- Claude: https://claude.ai/new |
|
9. Set up code for Github for HTML, Gradio, Streamlit examples |
|
10. Demonstrate and teach others. |
|
|
|
|
|
Github AI Example Repositories: |
|
1. HTML: https://github.com/AaronCWacker/AIExamples-8-24-HTML |
|
2. Gradio: https://github.com/AaronCWacker/AIExamples-8-24-Gradio |
|
3. Streamlit: https://github.com/AaronCWacker/AIExamples-8-24-Streamlit |
|
|
|
# Process: |
|
|
|
1. Stage your three github repos with code. |
|
2. In Azure Subscription search "App Services" |
|
3. Add each git to an app service and configure command line call to start it. |
|
|
|
Reference Papers and Spaces: |
|
|
|
STaR: Self-Taught Reasoner, Bootstrapping Reasoning With Reasoning: https://openreview.net/pdf?id=_3ELRdg2sgI |
|
From Google Gemini to OpenAI Q* (Q-Star): A Survey of Reshaping the Generative Artificial Intelligence (AI) Research Landscape: https://arxiv.org/pdf/2312.10868 |
|
AI Research Papers via ArXiV Gradio Space and API: - https://huggingface.co/spaces/awacke1/Arxiv-Paper-Search-And-QA-RAG-Pattern |
|
|
|
HTML5: 🌟Depth Anything w/ 🤗 Transformers.js🐆 - index.html - main app |
|
Gradio: https://huggingface.co/spaces/awacke1/Models-Datasets-Spaces-Search-Hub |
|
Streamlit: Transcript EDA NLTK - https://huggingface.co/spaces/awacke1/Transcript-EDA-NLTK |
|
|
|
|